RElSLER AND LAMED
4406. Yanofsky, C., Horn, V., Bonner, M.,and Stasiowski, S. (1971), Genetics 69, 409. Yguerabide, J., Epstein, H. F., and Stryer, L. (1970). J . Mol.
Biol. 51, 573. Zalkin, H. (1973), Adu. Enzymol. 38, 1. Zalkin, H., and Hwang, L. H . (1971), J . Biol. Chem. 246, 6899.
Interaction of Adipic Acid Dihydrazide Analogue of ATP with Myosin. Involvement of the Essential Sulfhydryl Groups? Emil Reisler*,l and Raphael Lamed5
ABSTRACT:
The hydrolysis by myosin of a soluble ATP analogue, adipic acid dihydrazide-ATP, is shown to proceed in a fashion similar to the hydrolysis of ATP by myosin modified at either of the twoessential sulfhydryl groups. In both systems, the Mg*+-activatedhydrolysis of the nucleotide is increased, whereas the EDTA-stimulated activity is inhibited. Blocking of one of the two essential sulfhydryl groups of myosin, S H I or SH2, leads to a complete inhibition of the analogue hydrolysis. The analogue is unable to expose the essential sulfhydryl group SH2 for modification by thiol reagents. Evidence is presented to show that the nucleotide derivative does not label irreversibly the protein. It is concluded that adipic acid
dihydrazide-ATP (ADH-ATP) need interact with only one of the two essential thiol sites of myosin. The hydrolysis of MgADH-ATP by myosin is inhibited by large excess of actin and does not result in contraction of actomyosin threads. MgADH-ATP is also a rather weak dissociating agent of the acto-heavy meromyosin complex. These properties of the ATP analogue are discussed in conjunction with the previous modification studies of myosin and the mechanism of ATP hydrolysis derived from them (Burke, M., Reisler, E., and Harrington, W. F. (1973), Proc. Natl. Acad. Sci. U.S.A.70, 3793; Reisler, E., Burke, M., and Harrington, W. F. (1974a), Biochemistry 13, 20 14).
R e c e n t l y it was suggested that the low rate of ATP cleavage in the resting state of muscle results from formation of a stable complex structure involving the two essential thiol sites (SHI and SH2) on each myosin head and MgATP (Burke et al., 1973; Reisler et al., 1974a). Activation was thought to occur by interaction of actin with myosin in the vicinity of one of the two essential thiol sites, thus breaking the stable complex and leading to a rapid dissociation of split products. In a way the action of actin was viewed as analogous to the chemical blocking of S H I sites of myosin; it was noted, in earlier studies, that such modification resulted in significant activation of myosin ATPase in the presence of Mg2+ ions (Sekine and Kielley, 1964) and the concomitant ioss of the ability of actin to activate this hydrolysis reaction (Silverman et al., 1972). I n terms of the proposed model (Burke et ai., 1973; Reisler et ai., 1974a), these two observations are corollary of the inability of the chemically modified SH1 site to participate in the formation of the stable complex structure with MgATP. It is assumed implicitly, in this context, that the above two effects of modification of S H I sites of myosin are linked and stem from the same mechanistic reason. If indeed true, such explanation would imply that the cyclic formation and opening of the inhibitory MgATP-myosin complex is an essential requirement of the contractile process. The last conclusion is supported within the qualifications set above, by the recent finding that the modification of SH1 sites
of myosin precludes contraction of actomyosin threads (Harrington et al., 1975). The above suggestions rely, however, upon a reasonable but unproved thesis, that chemical modification does not introduce structural changes, which may produce the observed alterations in the enzymatic properties of myosin. Thus, it would be desirable to find experimental conditions which lead to, or support, the same conclusions, without involving a chemical assault on the protein. In principle, such a goal could be achieved by employing an appropriate ATP analogue, which could not form the assumed cyclic complex with the two thiol sites. Such an analogue should display hydrolytic behavior (with native myosin) comparable to that of ATP with myosin modified at S H I or SH2 (i.e., activated MgATPase, inhibited EDTA(K+) ATPase, very small, or no actin activation, etc.). We require, thus, a limited structural modification of ATP, which would preclude the precise interaction with the two thiol sites, but would not impair the hydrolysis reaction. To our knowledge, none of the numerous ATP analogues previously tested with myosin meets the rather restrictive and specific requirements listed above (for a comprehensive review on ATP analogues and their interaction with myosin, see Yount, 1975). The comprehensive studies of Tonomura et al. (1967) have shown a tolerance of modification of the ribose ring of ATP for cleavage of phosphate and much less for contraction. In view of this, our attention focused on ribose modified ATP derivatives prepared by Lamed et al. (1973) and successfully employed in affinity chromatography of myosin fragments (Lamed and Oplatka, 1974; Oplatka et al., 1975; Muhlrad et al., 1975). We have chosen to study the interaction and hydrolysis by myosin of adipic acid dihydrazide-ATP (ADH-
~~~~~~
~
~
~
From the Polymer Department, The Weizmann Institute of Science, Rehovot, Israel. Receiued December 9, 1976. Present address: Department of Chemistry and Molecular Biology Institute, University of California, Los Angeles, California 90024. Present address: Miles Yeda, Kiriat Weizmann, Rehovot, Israel.
*
2532
BIOCHEMISTRY, VOL.
1 6 , N O . 1 1 , 1977
MYOSIN AND ADIPIC ACID DIHYDRAZIDE ATP ANALOGUE
ATP),’ since it was observed to be cleaved at a rapid rate in the presence of Mg2+ cations, displayed no activation by EDTA, and failed to cause shortening of myofibrils (Lamed et al., 1973). The results to be presented below suggest that ADH-ATP interacts with only one of the two essential thiol sites of myosin and that the ADH-ATP-myosin system has many characteristics in common with thiol modified myosin and ATP. We also show that the hydrolysis of ADH-ATP is inhibited by actin and does not result in contractions. Materials and Methods Glass-distilled water was used throughout and inorganic salts and reagents were of analytical grade. FDNB and MalNEt were the products of Eastman Kodak Co. (Rochester, N.Y.) and ATP of highest available purity was purchased from Sigma Chemical Co. (St. Louis, Mo.). The preparation of myosin and actin has been described elsewhere (Azuma and Watanabe, 1965; Lehrer and Kerwar, 1972). H M M and H M M S-I were obtained and purified using the previously described chromatographic procedure (Lamed and Oplatka, 1974). The preparation of Sepharose sebacic dihydrazide-ATP columns employed in purification of the proteins was somewhat modified. Sebacic dihydrazide was dissolved in 50% acetic acid (20 g/L) at room temperature and adjusted to pH 2.0 with HCI to ensure complete dissolution. Cyanogen bromide activated Sepharose 4B (0.5 volume) was added, and the mixture was stirred overnight at room temperature. Periodate oxidized ATP was then coupled to the resin as described earlier (Lamed and Oplatka, 1974). The ATP derivative, adipic acid dihydrazide-ATP, was prepared as described by Lamed et al. (1973). Selective modification of SH1 and SH2 groups of myosin with MalNEt or FDNB was carried out as reported by Reisler et al. (1 974a). Reduction of the dinitrophenylated myosin with sodium dithionite and removal of the dinitrophenyl group by thiolysis with 6-mercaptoethanol was performed as described before (Reisler et a]., 1974a). The reactions of myosin with MalNEt were carried out at 1% protein concentration in 0.5 M KC1-50 mM Tris-HC1, pH 7.9 at 5 OC. The reaction was terminated by a 20-fold dilution of the protein in 0.5 M KC1-Tris-HC1 solvent containing a 50-fold excess of dithiothreitol. The Ca2+- and EDTA-stimulated ATPases and ADHATPases, as well as MgADH-ATPase activities of myosin were measured at 37 “C employing the procedures of Kielley et al. ( 1 956) and Kielley and Bradley (1956).Temperature and pH dependence of myosin activities was monitored by a pHstat titration method in 50 mM KCI, 1 mM ATP or ADHATP, 1 mM Mg2+ or Ca2+ cations. In pH studies, the temperature was maintained at 25 OC;in temperature dependence measurements, the pH was kept at 7.6. Actin activation of the ATPase activity of myosin was measured on protein samples incubated for 5 min, at 37 OC, in a solution containing 50 mM KCI, 10 mM Tris-HCI, 5 mM Mg2+, and 2 mM ATP or ADH-ATP, at pH 7.6. Following the incubation, the protein was precipitated by 10% trichloroacetic acid and the supernatant obtained on centrifugation was assayed for P, by the method of Kielley et al. (1956).
’
Abbreviations used: ADH-ATP, adipic acid dihydrazide-ATP; AMP-PCP, a,@-methylene-ATP; AMP-PNP. P,y-imino-ATP; ATP(yS), adenosine 5’-0-(3-thiotriphosphate); FDNB, fluorodinitrobenzene; MalNEt, N-ethylmaleimide; HMM, heavy meromyosin; H M M S-I, heavy meromyosin subfragment I .
The dissociation of acto-HMM or acto-HMM S-I complexes by ATP or its analogue was followed by light scattering. The measurements (at an angle of 90’) were conducted at 25 “ C in a Fica (Paris, France) light scattering photometer employing octagonal cells as described by Eisenberg and Tomkins (1968). Protein concentrations in controls (uncomplexed H M M or actin) and in acto-HMM or acto-HMM S-I, were 0.07 mg/mL for H M M and H M M S-I, and 0.12-0.1 5 mg/mL for actin. To monitor the effect of ATP or ATP analogue on acto-HMM solutions, small volumes of the concentrated nucleotide solutions were added directly to the scattering cells containing the protein system. Prior to measurements, all solutions were filtered through 0.45-pm Millipore filters. In the presence of 1 mM MgATP, the acto-HMM complex is fully dissociated and the scattering of acto-HMM solution ( Z a c t o - ~ ~ reduces ~ ) to the sum of scatterings of its components, H M M and actin ( I H M M IaCt,,,). The observed drop in scattering ( A I = I a c t o - ~ ~lacto-HMM+nucleotide) ~ is a measure of the extent of dissociation of acto-HMM solutions relative to the dissociation in the presence of 1 mM MgATP.
+
Results The ability of myosin and myosin fragments to hydrolyze soluble ATP derivatives modified on the ribose moiety was demonstrated by Lamed et al. (1973). These authors noted that the derivatives, and particularly adipic acid dihydrazide-ATP, were cleaved at a rapid rate in the presence of Mg2+ and Ca2+ cations but could not be hydrolyzed in the absence of divalent cations, Le., in the presence of EDTA. Such hydrolytic behavior is rather similar to that of ATP with SHI-MalNEt or SH2-MalNEt modified myosin. Two possible explanations come to mind: either the derivative covalently binds to myosin and modifies its properties in a manner characteristic of a thiol reagent, or, because of altered geometry, it interacts reversibly with the protein without involving all the residues and structural transitions that ATP does. To examine these possibilities and their implications, we have studied the hydrolysis of ADH-ATP by myosin in greater detail. The various ADH-ATPase activities of myosin are summarized in Table I. The first row discloses that contrary to ATP, Mg2+ is a potent activator of ADH-ATP cleavage. In fact, MgADH-ATPase is comparable with MgATPase of SHI-MalNEt modified myosin and is manyfold higher than MgATPase of the native protein. Chelation of Mg2+ cations by EDTA activates the “normal” ATPase but almost completely inhibits the ADH-ATPase reaction in analogy to the effect of EDTA on cleavage of ATP by SHI-MalNEt (or SH2-MalNEt) myosin (row 2). In the presence of Ca2+cations, the rates of hydrolysis of ATP and its analogue are not significantly different (row 1). The divalent metal requirement for ADH-ATPase activity of myosin is shown in Figure 1. The requirement for Mg2+ is rather low since even at 10-6 M level of this metal and 1 mM ADH-ATP most of the activity is preserved. In contrast to this, relatively high Ca2+ concentrations are required for Ca2+-activated hydrolysis of the analogue and high Mg2+levels are necessary for ATP hydrolysis by SHI-MalNEt modified protein (dashed curve, from Burke et al., 1973). The shallow profile of ADH-ATPase dependence on Mg2+ level probably reflects rather tight binding of this metal to myosin. In the ATP-myosin system, such binding is necessary for Mg2+-induced inhibition of the hydrolysis reaction. An interesting situation arises when we examine the hydrolysis of ADH-ATP by SHI-MalNEt modified protein (row BIOCHEMISTRY, VOL.
16,
NO.
I 1, 1977
2533
TABLE I: Specific
Activitiesu of Native and Modified Myosins (pmol mg-l min-l). ATP EDTA(K+)
Myosin SH1-MalNEt myosin S H I - D N P myosin SH ]-reduced D N P myosin S H I - D N P myosin 0-mercaptoethanol
+
2.6 0.15 0.42 (16) 1.95 (75) 1.82 (70)
ADH-ATP
MgZ+
EDTA(K+)
0.015c 0.12c
0.05 0.01
Ca2+ 0.95 2.3 3.40 (350) 3.30 (350) 1.05 (110)
Ca*+ 1.2 0.08 0.23 ( 1 9) 0.66 (55) 0.84 (70)
Mg’S
0.45 -0 0.09 (70) 0.38 (85) 0.29 (65)
Activity measurements were carried out at 37 OC, pH 7.6, in 0.05 M KCI for Ca-ATPase and in 0.8 M KCI for EDTA-ATPase (Kielley et al., 1956; Kielley and Bradley, 1956), unless otherwise specified. The numbers in parentheses refer to relative activities. ‘ Measurement, carried out at 25 ‘C, pH 7.6, in 0.05 M KCI employing the pH-stat titration method. (I
-__
T A B L E 1 1 : Specific Activities of Myosin Incubated with MgADH
A T P (pmol mg-’ min-l).”
Step Step I Step I 1
- l o g Mt’
FIGURE 1: Relative ADH-ATPase activity of myosin as a function of the concentration of added divalent cations: ( A ) Ca2+;( 0 )Mgz+. The dashed line shows the relative ATPase activity of SHI-MalNEt modified myosin as a function of Mg2+ ion concentration (from Burke et al., 1973). In the absence of Mz+ cations, ;.e., in the presence of EDTA, ADH-ATPase activity is inhibited.
2); practically all of the analogue activities are lost. Clearly, retention of only one of the two essential thiol groups of myosin is not sufficient for expression of ADH-ATPase activities, although it is adequate for the cleavage of ATP itself. Involvement of the thiol sites in the reaction of myosin with ADH-ATP can also be tested through dinitrophenylation of S H I groups and subsequent manipulation of the attached label (Reisler et a]., 1974a). As shown in Table I, modification of S H I groups with FDNB induces a similar, although less dramatic, effect as labeling with MalNEt (row 3). Reduction of the DNP label with sodium dithionite, Le., formation of a nucleophilic center at the SH1 site leads to a partial regeneration of Mg and Ca ADH-ATPases (row 4).Similar recovery of analogue activities is achieved on reversal of DNP binding by incubation of the modified protein with P-mercaptoethanol (row 5). Thus, in analogy with the ATP-myosin system (Reisler et al., 1974a), SHI sites are seemingly implicated in the interaction with the metal-analogue complexes. It should be mentioned, however, that also SH2-MalNEt modified myosin, which could cleave CaATP and MgATP, did not hydrolyze ADH-ATP. It could be that, because of the proximity of the essential thiol groups spaced at about 10 A apart (Reisler et al., 1974b; Haugland, 1975), blocking of one of them sterically prevents the interaction with the other site. To check whether the cleavage pattern of ADH-ATP does not derive from an irreversible attachment of the analogue to the vicinity of S H I or SH2 site, we have examined a composite ADH-ATP and ATP system. In the first step, myosin assay system was incubated in the presence of 1 mM ADH-ATP and 1 mM Mg, Ca, or EDTA. Following this incubation half of the
2534
BIOCHEMISTRY. VOL.
16,
NO.
1 1 , 1977
Present in the incubation mixture ADH-ATP, 1 mM ADH-ATP ( 1 mM): ATP ( 5 m M )
Mg2+
Ca2+
EDTA
0.45
I
__0
02
04
06
C
08
IO
i -
2
nucleotide (mM1
7: Dissociation of acto-HMM complex induced by MgATP (o), MgADH-ATP (O), and EDTA-ATP ( A ) plotted against their respective concentrations. The dissociation was followed by light scattering as described in Materials and Methods. One hundred percent dissociation refers to full dissociation of acto-HMM into actin and HMM i n the presence of 1 mM MgATP. FIGURE
IO~/T(K-~
FIGURE 5: Arrhenius plots of log of specific activities of myosin against reciprocal temperatures. A, A , and 0 represent respectively CaATPase, CaADH-ATPase, and MgADH-ATPases of native myosin. The dashed line (V) shows the CaATPase of SHI-MalNEtmodified myosin.
Ca ADH-ATP I
60
8.0
70
9.0
PH
FIGURE 6: Effect of pH on relative ATPase ( 0 ,A) and ADH-ATPase activities ( 0 , A ) of myosin at 25 O C . (a) Ca2+-stirnulated and (b) Mg2+-stirnulated activities.
their behavior as a function of pH. Figure 6 demonstrates the difference between ADH-ATP and ATP activities of myosin as a function of hydrogen ion concentration. The ADH-ATP pattern in this figure is closely similar to that observed with ATP and SHI-MalNEt modified myosin (Sekine and Kielley, 1964), and that of myosin-ATP system at low temperatures (Watterson et al., 1975). It is tempting to relate again the changed activity-pH profile of the last three cases with the absence of the rate-limiting complex, M**ADP.P,, in these systems. The results discussed so far raise the question of the interaction of the analogue with the actomyosin system. The observation of Lamed et a]. (1973), that ATP derivatives of this type did not cause contraction of myofibrils, suggested that their hydrolysis is abortive in the mechanistic sense. We have verified this result by following the shortening of actomyosin threads employing previously described procedures (Harrington et al., 1975). MgADH-ATP was unable to cause shortening of threads, whereas MgATP brought about their rapid shrinkage. Activity assays carried out on actomyosin systems used for preparation of such threads disclosed a high
2536
BIOCHEMISTRY. VOL.
16, NO. 1 1 , 1977
rate of ADH-ATP hydrolysis, only slightly lower than that of ATP in the same system. Increasing the ratio of actin to myosin has an inhibitory effect on the rate of cleavage of ADH-ATP. At 1:1 actin/ myosin ratios (w/w) the inhibition of ADH-ATPase reaches about 75% of the initial activity value and exceeds 90% at higher actin/myosin ratios. The same behavior was also observed when myosin was replaced by HMM or HMM S-I. To rule out the possibility that actin promotes an irreversible inhibition of myosin by ADH-ATP, we have performed tests similar to those presented in Table 11. Maximally inhibited MgADH-ATP actomyosin was supplemented with an excess of MgATP, whereupon rapid liberation of P, ensued, thus confirming that myosin was not covalently modified. Similar inhibition of myosin ATPase by actin was noted before to occur in the presence of EDTA (Baron et al., 1966) and to a smaller extent in the presence of Ca2+ cations (Nanninga, 1964). It seems that in all these cases the inhibition by actin reflects a decreased affinity of the nucleotide for myosin and its consequent displacement by high concentrations of actin. A competition between actin and ATP also prevails in the presence of Mg2+cations; yet in the latter case the binding of nucleotide to myosin is strong enough to dissociate the actomyosin complex, even at high actin/myosin ratios and thus allow for the cyclic interaction of MgATP and actin with myosin and stimulation of the hydrolysis rate. Figure 7 demonstrates that indeed MgADH-ATP and EDTA-ATP have apparently lower affinity for myosin than MgATP. Very low concentrations of MgATP (10 pM) are sufficient to induce full dissociation of acto-HMM, or acto-SHI-MalNEt-HMM,2 as detected by a drop in the light scattering of the solution to a value of combined scatterings of the isolated components, H M M and actin. Many-fold higher levels of MgADH-ATP (about 1 mM) or EDTA-ATP are required to achieve the same effect (Figure 7). It appears from the data of Figure 7, which were also reproduced for the acto-subfragment-I system. that at high actin/myosin ratios mM levels of MgADH-ATP or EDTA-ATP may be too small to displace actin. The net result of this is the suppression of hydrolysis of the appropriate nucleotide. It should be stressed, however, that the failure of
* At such low ATP concentrations, dissociation of acto-HMM o r acto-SH1-MalNEt-HMM is succeeded by the immediate recombination of actin and H M M following the hydrolysis of ATP. Note that the hydrolysis of MgATP by SH,-MalNEt myosin is not inhibited by actin. Thus, both with respect to the dissociation of acto-HMM and inhibition of hydrolysis by actin the myosin--ATP analogue system differs from the thiol modified myosin-ATP system.
MYOSIN A N D ADIPIC ACID DIHYDRAZIDE ATP ANALOGUE
rived from them. The results with ADH-ATP confirm that inhibition of phosphate liberation by Mg2+ cations requires interaction of the phosphate nucleotide with the two essential thiol sites of myosin of striated muscle. The fact that the hydrolysis of MgADH-ATP by myosin is not activated by actin nor does it result in contraction in spite of a relatively high turnover rate is instructive. We correlate it with the inability of the analogue to interact simultaneously Discussion with the two essential sulfhydryl sites of myosin. Seemingly, such an interaction is necessary to produce a cyclic conforThe mechanism of ATP hydrolysis by myosin may be indimational change on the myosin head without which contraction rectly studied by monitoring the alterations in the enzymatic cannot occur. Modification studies suggested this might be the reaction of myosin following specific modifications of funccase (Harrington et al., 1975), though they could not obviate tional residues. In fact, such an approach has led to formulation the possibility that the loss of the contractile function and the of a model, which attributes the low rate of ATP cleavage in activation of MgATPase of myosin through blocking of S H I a resting muscle to the formation of a stable complex structure groups may be only casually related. In this sense the results with the participation of MgATP and the two essential thiol of ADH-ATP add credence to our previous proposal that sites (Burke et al., 1973; Reisler et al., 1974a). This model is contraction of striated muscle requires a cyclic opening and reminiscent of the “product lock-in” explanation suggested by closing of a stable complex formed with the participation of Werber et al. (1972). Wider use of the modification approach MgATP and the two sulfhydryl sites, S H I and SH2. is limited not only for practical and technical reasons but also The stable complex, which we equate with the rate-limiting because of the realization that modifications change the protein complex in the hydrolytic cycle of myosin, M**ADP.Pi, may and the interpretation of results accordingly requires caution. be detected by a number of methods, most conveniently by The conclusions derived from such investigations can be, fluorescence measurements (Werber et a]., 1972). Whenever however, reinforced if supported by experiments in which a M**ADP.Pi cannot be detected, contraction or activation of modified substrate (analogue) rather than a modified protein has been used. The most frequently used ATP analogues, myosin MgATPase by actin will probably not follow. Thus, for example, by fluorescence criterion, MalNEt modified ATP(yS), AMP-PCP, and AMP-PNP, display either a HMM does not display the existence of M**ADP.Pi (Werber changed hydrolysis reaction or their cleavage by myosin is et al., 1972) and its capacity to hydrolyze ATP is little affected totally inhibited. Clearly, the analogues generate different by actin (Silverman et al., 1972). The ATP analogue examined steady-state complexes from that obtained with ATP, a feature in this work neither supports contraction nor does it enhance of great advantage for structural analysis of intermediate states the fluorescence of myosin. The lack of the fluorescence effect of actively contracting muscle (Goody et al., 1975). However, with MgADH-ATP suggests, in line with the kinetic evidence, because the analogues are different from ATP, they do not provide the final clue to the understanding of the mechanism that M**ADP.Pi is not formed, or is not predominant. One has to note, however, that detection of the optical properties asof ATP hydrolysis by myosin. Obviously, they also cannot cribed to the M**ADP.Pi state should not be equated with the probe the chemical events which underlie the mechanochemcontractile function, though their absence does imply the loss ical cycle. of this function. Optical (Werber et al., 1972) and chemical In the particular case of myosin, the model derived from (Watterson et al., 1975) probes attest that M**ADP.Pi is thiol modification studies suggests that an ATP analogue caformed with CaATP as well; yet no contraction follows. Evipable of interacting with but one of the two essential sulfhydryl dently, the CaATP and MgATP complexes with myosin differ groups of the protein should be an attractive one to test. From structurally in spite of giving rise to similar perturbation of previous work of Lamed et al. (1973), it appeared that spectral properties of the protein (see also Reisler et a]., ADH-ATP might serve this purpose. As documented in Table I the hydrolysis of ADH-ATP by myosin resembles that of 1974a). The characteristics of the M**ADP-Pi state may be exATP by SHI-MalNEt or SH2-MalNEt myosin; Mg2+cations activate the hydrolysis of ADH-ATP, whereas EDTA inhibits tended to explain the pH and the temperature dependence of it. Blocking of either S H I or SH2 groups of myosin leads to myosin ATPase and to correlate the transition in their profiles almost complete inhibition of ADH-ATPase (Table I) and, with changes in population of the intermediates of the hydrolytic cycle. Both Mg2+ and Ca2+ cations induce at room what perhaps is more significant, the analogue does not expose the SH2 group for modification by thiol reagents (Figures 3 temperature a double maximum pH profile of ATPase activity. Moreover, CaATPase and MgATPase display a biphasic and 4) as does ATP. Since we have verified that, under the experimental conditions employed in this study, the analogue Arrhenius plot with activation energies about 20-30 kJ/mol. does not covalently label the protein, the above results lead to When conditions of the enzymatic assay are changed, in a way the conclusion that ADH-ATP interacts with only one eswhich depletes the system from the M**ADP-Pi state, the pH sential thiol site at a time. Similar results were also obtained dependence of ATPase is transformed into a single maximum with pimelic and sebacic acid dihydrazide derivatives of profile and the activation energies are significantly increased. ATP. Interestingly, such change can be achieved simply by lowering The proximity of SHI and SH2 groups, of the order of 10 .fithe temperature of the assay to 0 O C (Watterson et al., 1975), (Reisler et al., 1974b), precludes definitive resolution of the which, as known from other studies, decreases the concentraidentity of the participating thiol site since the interaction of tion of M**ADP-Pi in the system (Bagshaw and Trentham, ADH-ATP with S H I or SH2 may be sterically hindered by 1974; Seidel, 1975; Harrington et al., 1975). A similar shift the label on the other group. That similar hydrolytic behavior of the pH profile of ATPases and increase in the activation could be observed for the ATP-thiol modified myosin system energy of the reaction (to about 90-100 kJ/mol) can be oband the analogue-unmodified myosin system lends support to served as a result of modification of the S H groups of myosin the utility of the modification studies and the conclusions de(Sekine and Kielley, 1964). The Arrhenius plots of the acMgADH-ATP to elicit contraction of actomyosin threads is not due to an unfavorable competition for myosin between actin and the analogue since at equimolar ratios of actin to myosin their complex is almost completely dissociated by 1 mM MgADH-ATP and the hydrolysis of the latter is practically uninhibited. Also, increasing the nucleotide concentration about tenfold does not lead to contraction of the threads.
BIOCHEMISTRY, VOL.
16, N O . 11, 1977
2537
REISLER
tivities of the thiol modified myosin also become linear. Hydrolysis of ADH-ATP, which by kinetic, spectral, and chemical evidence does not yield the M**ADP.Pi complex with myosin, falls into the same pattern of temperature and pH dependence (Figures 5 and 6). The high activation energy of the analogue activities of native myosin and ATPase activities of modified myosin can be compared with the large temperature dependence of the product dissociation step (MADP e M ADP). Bagshaw (1975) reports for this step an activation energy of about 130 kJ/mol, whereas the rate-limiting isomerization (M**ADP.Pi e M*ADP-Pi) is according to them practically temperature independent. Thus, the results presented in Figure 5 are consistent with other lines of evidence that M**ADP-Pi is not formed (or is not predominant) with ADH-ATP. In view of the analogies drawn above, it would be interesting to extend the conformational, kinetic, and contractile studies on the ATP-myosin system into the range of low temperatures, below 0 "C. The attractive possibility in such a study is that one can reach conditions where M**ADP.P, will be almost completely depleted from the system. The recent work of Seidel (1975) suggests that, though at 0 "C the rate-limiting complex still exists, lowering the temperature should bring us to the desired goal. Under such conditions, of no M**ADP.Pi, one may be able to characterize the other intermediates of the hydrolytic cycle and hopefully to establish, by a different approach, that contraction requires the cyclic formation and disruption of the Mg2+ induced M**ADP-Pi conformation.
+
Acknowledgment Thanks are due to Dr. Robert Josephs for his valuable comments. References Azuma, N., and Watanabe, S. (1965), J . Biol. Chem. 240, 3847. Bagshaw, C. R. (1975), FEBS Lett. 58, 197. Bagshaw, C. R., and Trentham, D. R. (1974), Biochem. J . 141, 331. Baron, S., Eisenberg, E., and Moos, C. (1966), Science 151, 1541. Burke, M., Reisler, E., and Harrington, W. F. (1973), Proc. Natl. Acad. Sci. U.S.A. 70, 3793. Eisenberg, H., and Tomkins, G. M. (1968), J. Mol. Biol. 31,
2538
B I O C H E M I S T R Y ,V O L .
16,
NO.
11, 1977
A N D LAMtlI
37. Goody, R. S., Holmes, K. C., Mannherz, H. G., BarringtonLeigh, J., and Rosenbaum, G. (1975), Biophys. J . 15, 687. Harrington, W. F., Reisler, E., and Burke, M. (l975), J . Supramol. Struct. 3, 112. Haugland, R. P. (1975), J . Supramol. Struct. 3, 338. Kielley, W. W., and Bradley, L. B. (1956), J . Biol. Chem. 218, 653. Kielley, W. W., Kalckar, H. M., and Bradley, L. B. (1956). J. Biol. Chem. 219, 95. Lamed, R., Levin, Y., and Oplatka, A. (1973), Biochim. Biophys. Acta 305, 163. Lamed, R., and Oplatka, A. (1974), Biochemistry 13. 3137. Lamed, R., Oplatka, A., and Reisler, E. (1976), Biochim. Biophys. Acta 427, 688. Lehrer, S. S., and Kerwar, G. ( 1 972), Biochemistry I I , 1211. Muhlrad, A., Lamed, R., and Oplatka, A. (1975), J . Biol. Chem. 250, 175. Nanninga, L. B. (1964), Biochim. Biophys. Acta 82, 507. Oplatka, A., Lamed, R., and Muhlrad, A . ( 1 975), Biochim. Biophys. Acta 385, 20. Reisler, E., Burke, M., and Harrington, W. F. (1974a), Biochemistry 13. 2014. Reisler, E., Burke, M., Himmelfarb, S.,and Harrington, W. F. (1974b), Biochemistry 13, 3827. Seidel, J. C. ( 1 975), J. Biol. Chem. 250, 568 1. Sekine, T., and Kielley, W. W. (1 964), Biochim. Biophys. Acta 81, 336. Sekine, T., and Yamaguchi, M. (1963), J . Biochem. (Tokyo) 54, 196. Silverman, R., Eisenberg, E., and Kielley, W. W. (1972), Nature (London) 240, 207. Tonomura, Y., Imamura, K., Ikehara, M., Uno, H., and Harada, F. (1967), J . Biochem. (Tokyo) 6 1 , 460. Watterson, J. G., Schaub, M. C., Locher, R., DiPierri, S., and Kutzer, M. (1975), Eur. J . Biochem. 56, 79. Werber, M. M. (l976), private communication. Werber, M. M., Szent-Gyorgyi, A . G., and Fasman, G. D. (1972), Biochemistry 11, 2872. Yamaguchi, M., and Sekine, T. ( 1 966), J . Biochem. (Tokyo) 59, 24. Yount, R. G. (1975), Adv. Enzymol. 43, 1.
